Home | News | Documentation | Download

Upper limit on particle width using ProfileLikelihoodCalculator?

Hello experts,

I am trying to set an upper limit on the width of a particle. The mass resolution is already determined from MC. The full mass distribution is fitted with the resolution function convoluted with two Gaussians. The plot of the negative profile likelihood ratio is shown below. The result I got from PLC is an interval like 12 +- 15 MeV. How do I turn this into an upper limit? All the examples I saw in the tutorials seen to be based on counting experiments.

Thanks
image

I’m not sure to understand the issue… Maybe @moneta can help?

Hi,

If you want to set an upper limit, then you should use the HypoTestInverCalculator based on the AsymptoticCalculator and possibly with the CLs procedure, that is typically recommended.

Given a RooFit model, you can use the tutorial StandardHypoTestInvDemo.C for this, see

https://root.cern.ch/doc/master/StandardHypoTestInvDemo_8C.html

Lorenzo